- 1. How are the mobile devices models and mobile devices producers rankings presented?
Please be informed that the method of identifying mobile devices in gemiusTraffic was modified as of 6.05.2010. Since some additional parameters provided by user browsers can now be obtained, the identification of mobile devices in the research has increased. This increase was mainly due to the fact that the identification of the mobile devices with the Opera Mini browser installed is possible. Due to the fact that the report for the week between 3.05.2010 - 9.05.2010 covers the data obtained with the use of two different methods, the data is not comparable with either the earlier data or the data from the subsequent weeks. Due to this situation we have decided not to present them on the website and to present the following weeks since 10.05.2010. All persons who wish to obtain archive ranking statistics, basing on data collected from the User-Agent header solely, are asked to contact us.
- 2. How do we present the ranking of "Color settings"?
Starting on March 16, 2011 the ranking of "Color settings" will be presented as follows:
Deep color as a total result for 64, 48, 40, 30-bit color settings,
True color as a total result for 32, 24-bit color settings,
High color as a total result for 18, 16, 15, 12-bit color settings,
Indexed color as a total result for 8, 7, 6, 5, 4, 3, 2-bit color settings,
Monochromatic color for 1-bit color settings.
The above presentation increases the transparency and the usability of gemiusRanking results.
- 3. How to interpret the countries and regions rankings?
IP address are matched to a particular country or region with a method specifically designed for this purpose by Gemius SA. The regions ranking is based on a percentage share of page views made from a such identified area on the researched Polish websites, hence the data does not necessarily match declarative data obtained in the studied country. The ranking of countries presents the mean share of page views made from the above-listed countries on an average Polish website.
- 4. What´s new on the web site?
We have modified the data presentation threshold, added some metrics and changed the layout – these are only some of the changes we have introduced to gemiusRanking PL, a site dedicated for those interested in being up-to-date with the trends prevailing on the Polish market.
Form of data presentation
The new gemiusRanking PL version features data concerning page views and visits made by internet users visiting Polish sites inclued in gemiusTraffic research, as well as ranking of countries from which foreign visitors connect to Polish web sites. The data available on gemiusRanking PL concerns only the local internet users.
By default, the tables present statistics for the last 3 weeks. A novelty is the possibility to select and compare data for any 3 weeks starting from 1 January 2007. When switching weeks, the names in the table are sorted by the data from the first column.
The rankings are presented as percentage by default. You may change the way data is presented into absolute values in case of stats calculated basing on page views. To do that, just tick the "Show absolute values” checkbox in the bottom left corner of the table.
There has also been a change to the "ranking week” start day definition. Previously, the "ranking week” started on Tuesday and ended on Monday. Now the seven-day statistics are calculated between Monday and Sunday.
The way statistics are presented in the rankings has changed as well. For more information on this subject go to "What's the value of data presentation threshold?”.
New stats
The refreshed gemiusRanking PL offers presentation of data for two new statistics – Flash Player versions and mobile devices used by internet users visiting Polish web sites.
- 5. What´s the value of data presentation threshold?
In the following rankings: web browsers (groups, details, engines), operating systems, mobile devices (models, producers), Flash Player, resolutions, color settings, directories, search (domains, engines) and countries are presented statistics which in a current week attained the percentage value equal to or exceeding 0.1%. In the event that a particular statistic reaches the percentage value lower than 0.1% in a current week, a yearly history of such statistic is calculated. The statistic is presented if within the last 52 weeks it exceeded the value of 1% over 3 weeks at least.
Data included in the mobile devices rankings (models, producers) is verified in a manner similar to the one described above, but with the difference that a 26-week statistics history is taken into account.
Rankings of regions are presented if they reached or exceeded the value of 0.01% in the current week. If a given statistic has not reached 0.01%, the half-year history of the very statistic is checked. Consequently, a statistic is presented if within previous 26 weeks it exceeded the 0.1% value over a period of at least 3 weeks.
Statistics in web browsers – details are presented under the assumption that the 0.1% value of data presentation threshold is established for the percentage values for the entire group of statistics. This means that, e.g. the share of Firefox 1.5 in considered for the Firefox group, and not its share in all browsers.
In operating systems – families and cookie acceptance rate rankings, the statistics are presented regardless of their percentage or absolute value.
- 6. What is meant by "page view", "visit" and "visitor (cookie)"? What´s the difference between the number of page views and visits?
Page view is an event consisting in an internet user viewing a web page. The gemiusTraffic study counts the page views thanks to the counting scripts inserted in the monitored page's source code.
Visit is a sequence of page views made by one visitor (cookie) within one web site. The time elapsed between subsequent views cannot exceed 30 minutes. If an internet user makes another view after the 30-minute limit, the view will be assigned to the next visit. The number of visits statistic presents the number of visits involving a particular web page. If, within one visit, an internet user visited a particular page 3 times, he/she increased the number of this page's views by 3, while the number of visits involving this page increased by 1.
Visitor (cookie) is a person using the WWW, identified with the use of cookie files assigned to the profile of the used browser. Technically speaking, this is a unique cookie.
One internet user can make any number of visits. Each visit comprises a number of pages. If, within a week, a visitor (cookie) made 5 views of a page, this translates into an increase in the number of this page's views by 5, whereas the number of users (cookies) for this page in this particular week has increased by 1.
A visitor (cookie) does not translate into a real person using the internet. If a person uses more than one browser on one computer, than he/she will be identified as an individual visitor (cookie) with each of such browsers. If a user connects from several different computers, e.g. at school, work, home, etc., in each of these locations he/she will be seen as a separate visitor (cookie).
- 7. What is "the percentage number of visits"?
For the sake of explanation, let's take the example of search engines. If there have been 20 visits referred by search engines to web site A during a week:
- 5 visits referred by the search engine X
- 15 – from the search engine Y
than, thanks to this web site, search engine X has attained 25%, and search engine Y –75%.
By contrast, if we analyse a web site B, which has 2000 referrals from search engines a week, and in a week you get: 500 visits referred from search engine X and 1500 visits from search engine Y, the percentage will remain the same. Website B will also give search engine X – 25%, and search engine Y – 75%.
Then, an average of the percentage of visits is calculated for each search engine. This average is the search engine’s result in the ranking.
- 8. Does a ranking based on this methodology reflect the actual usage of search engines?
The analysis on groups of web sites with varied traffic referred by search engines shows that the intensity of traffic referred by search engines has no significant influence on the search engines’ ranking results. The ranking is calculated on the basis of the percentage of visits referred by search engines to sites monitored by gemiusTraffic. Therefore, direct measurement of traffic on the search engines’ web pages is not of concern here. In line with the methodology described above, all web sites – these of heavy and of little traffic – have the same basic value to be distributed, i.e. 100%. So, we can assume that the ranking is a combination of two elements:
- internet traffic referred by search engines in a particular country,
- the reach of search engines, which is the number of web sites to which search engines refer users.
- 9. Why is the percentage of visits referred by search engines a basis for calculating the ranking?
Such methodology secures the ranking results against distortions which may be caused by web sites with heavy web traffic. If we presented a ranking based on the absolute number of visits referred by search engines, than putting a web site of heavy traffic at the top of a particular search engine’s list (for example as a sponsored link) could improve the position of this search engine in the ranking. Such threat was eradicated in the new ranking by providing web sites of heavy traffic with the same “say” as web sites of little traffic.
- 10. Is the search engine ranking based on the currently applied methodology susceptible to distortions?
There is a theoretical possibility of distortion. It could take place if a search engine operator deliberately ranked gemiusTraffic-monitored sites higher. This is, however, highly unlikely, as the quality and objectivism are the top priorities for the search engines’ operators (and favoring web sites only because of the fact that they are using the gemiusTraffic research contradicts this priority).
- 11. What is the role of cookies?
Cookie is a file provided by a server for a browser during a page view. Information contained in the file is sent back (and potentially modified) to the server each time the browser requests a page. The main purpose of cookies is to identify visitors returning to a web site.
Some people regard cookies as an invasion of privacy and block cookies in their browsers.
Most services which require identification of the internet user who views a page, such as online shops or banks, use cookie files. If it were not for the cookies, a person wanting to use an online bank account would have to provide the password on each opened page. Moreover, cookies allow for user customisation of web pages.
Should an internet user regard cookies as a threat for his/her privacy, or for any other reason does not wish to have them assigned, the cookie files can be disabled (e.g. in Firefox 3 Tools > Options...> Privacy > Enable cookies). Cookies are not accepted by less than 2% of internet users – current data is available in the Rankings tab.
- 12. Can the information collected be gemiusTraffic by used to identify internet users?
The gemiusTraffic application does not collect any personal data of internet users.
What is gathered in the study is only the information on web browser and computer of the user visiting a monitored site. This allows to determine what operating system, screen resolution, color depth, browser, etc., the web site visitor is using, and to determine where the user has connected from (basing on the IP). It is impossible to identify anyone personally using this data.
Visitors (cookies), defined during the research do not represent actual individuals using the internet, but the browsers used to visit a web site. This is a consequence of the fact that an internet user is identified by means of cookies assigned by a server. If an individual uses several browsers, that individual will be seen as a different visitor (cookie) for each of the browsers separately.
The fact that gemiusTraffic does not collect personally identifiable information has been verified through an independent legal counselor expertise. The term "personally identifiable information" is defined in Article 2 of the Act of 29 August 1997 on personal data protection (Journal of Laws No. 133 of 1997, as amended).
Personally identifiable information is information about a physical person whose identity can be directly or indirectly determined, specifically through the use of a unique identification number or any other data describing the physical, physiological, mental, cultural or social characteristics. gemiusTraffic does not gather any of such data.
- 13. How to download gemiusRanking PL data?
Ranking data may be downloaded in XML and PDF formats.
Data for last week may be downloaded by clicking XML and PDF icons which may be found in the bottom menu of gemiusRanking PL web site or through:
- http://www.ranking.pl/en/rankings.xml (XML format),
- http://www.ranking.pl/en/rankings.pdf (PDF format).
The statistics for the last three weeks may be downloaded from:
- http://www.ranking.pl/en/previous.rankings.xml (XML format),
- http://www.ranking.pl/en/previous.rankings.pdf (PDF format).
If you are interested in different data, please contact us by e-mail or through our contact form.
- 14. Where do I find archived data?
The founder of this web site – Gemius SA – modified the calculation method for operating systems, screen resolution, color depth, browser, provinces and countries rankings in 2007. The change involved replacing the visitors count with a method based on the number of page views.
You can get the archived visitor (cookie) – based statistics for 2001-2007 at www.old.ranking.pl.
Data based on the number of page views, archived since January 2007, can be accessed by selecting the desired week in the table in the current gemiusRanking PL web site.